DOCKGROUND resource for protein recognition studies

Keeley W. Collins,Matthew M. Copeland,Ian Kotthoff,Amar Singh,Petras J. Kundrotas,Ilya A. Vakser

DOI: https://doi.org/10.1002/pro.4481

IF: 8

2022-10-26

Protein Science

Abstract:Structural information of protein‐protein interactions is essential for characterization of life processes at the molecular level. While a small fraction of known protein interactions has experimentally determined structures, computational modeling of protein complexes (protein docking) has to fill the gap. The Dockground resource (http://dockground.compbio.ku.edu) provides a collection of datasets for the development and testing of protein docking techniques. Currently, Dockground contains datasets for the bound and the unbound (experimentally determined and simulated) protein structures, model‐model complexes, docking decoys of experimentally determined and modeled proteins, and templates for comparative docking. The Dockground bound proteins dataset is a core set, from which other Dockground datasets are generated. It is devised as a relational PostgreSQL database containing information on experimentally determined protein‐protein complexes. This report on the Dockground resource describes current status of the datasets, new automated update procedures and further development of the core datasets. We also present a new Dockground interactive web interface, which allows search by various parameters, such as release date, multimeric state, complex type, structure resolution, etc., visualization of the search results with a number of customizable parameters, as well as downloadable datasets with pre‐defined levels of sequence and structure redundancy. This article is protected by copyright. All rights reserved.

biochemistry & molecular biology